THE BOOK--Playing The Percentages In Baseball

Whatever Hudson gets, that’s what Ellis would probably have gotten, in an open market.

I think I understand why you write this, but do you really think that GMs see these players the same way the readers here do? I’m not putting one group or another down, but while I think that people here have a decent idea of what the players are worth ont he field, there are a lot of posts made here that indicate that GMs don’t know what players are worth, particularly players like Ellis who have much of their value tied up in excellent defense.

I agree that Ellis had nothing to lose by filing for free agency, he could have gotten more, etc. I guess I’m asking you to explain why you think offseason decision makers would see Ellis and Hudson as roughly equal, given the numerous posts on here about how most of them don’t really understand how to evaluate fielding, positional value, etc.

They may or may not know how to evaluate fielding, but Hudson and Ellis would be considered equivalent in terms of their components.  Even if Hudson will get a contract that is less than what a 2.95 WAR all-hit no-field guy born in 1977 would get (just to give the player a face, say Travis Hafner… I don’t know what his WAR is), that’s *not* the comparison point.

I’m strictly focusing on Ellis v Hudson.  Whatever the GMs value Hudson, that’s how much they’ll value Ellis, because they are twins.

"Whatever the GMs value Hudson, that’s how much they’ll value Ellis, because they are twins.”

I don’t think that necessarily follows. Hudson hit .305 this year while Ellis hit .233. Don’t you think that difference will have some effect on how GMs value them (regardless of whether it should or not)?

I apologize if I’m being thickheaded about this, but I’m not sure I get it yet or if I’ve put my question the right way.

We see Ellis and Hudson as twins who have the same market value. But are GMs? The players may have identical WAR in our mind, but they get there different ways—Ellis has better defense over the last three years, Hudson better offense. If, for example, the GMs don’t know how to evaluate fielding and see them as equal fielders with Hudson having better offense, they’ll see Hudson as the better player worth a bigger contract, right?

I’m not trying to be difficult, and I won’t badger you, I think I’m just missing the point at the moment.

Franklin/3: all we need is ONE general manager to see Ellis’ value for what it is.  So, no, I don’t think that it plays a part in it. 

***

Devil/4: You are right that Hudson and Ellis don’t get there in exactly the same way.  You are suggesting that readers of this blog see Hudson and Ellis as overall twins, but that GMs are going to call their fielding a wash, and award Hudson the larger WAR.  Sure, that’s possible.  What are we talking about here?  0.5 WAR?  That’d be 2MM a year.  So, if Hudson signs a 4/48 deal, then that means Ellis would be worth a 4/36 deal, under the idea that GMs undervalued Ellis by 0.5 wins, relative to the fans view.

Let’s see how it plays out…

I expect Hudson to have much more value than Ellis despite their similarities.  Hudson is a scrappy player, and GM’s like that.  Ellis is probably viewed as sort of lazy, maybe a malingerer always asking out of lineups and such, a true natural athlete who’s a good player, but hasn’t come close to reaching his potential.

TT/5: OK, that’s fine. I agree that Ellis signed for 50 cents on the dollar of his value (at best, actually, or would it be at worst). My question was more about the “twinning” in the eyes of the GMs.

Like I said before, I follow the Royals, where the alleged “respected baseball man” GM signed Jose Guillen to 3/36 when it would have been a stretch (in my opinion) to go 3/22. Of course, it turns out that even the latter amount would have been about $22 million off, given Guillen’s heroic sub-replacement performance this year.

I’m pretty sure Dayton Moore would look at Ellis and Hudson as having the same value(let’s imagine he thinks in terms of “Wins Above Replacement,” just to be crazy):

“Let’s see, I have them each at 3 WAR. Hudson seems more athletic, so that’s +0.5 for him (3.5). But Ellis is a lot grittier, so that’s +1 for him (4.0). Ellis is not from the South, so that’s -0.5 (3.5). Hudson IS from the south (Like my boy Frenchy!), so that’s +0.5 (4.0). However, I don’t like that “O-Dawg” nickname, so that’s -0.5 (3.5). Wow, a dead heat. GMing the Royals is hard. Better read that passage from Philippians again…

...OK, even at 3.5. Hudson played in the National League, and I can tell you from my years in Atlanta, with all the bunting and double switches and stuff, it’s a lot harder over there, so that’s +0.5 for Hudson (4.0). But he also played for the jerk in Toronto who used to work with Billy Beane, so -0.5 for Hudson. Dang, still even at 3.5. Well, I know this much—I didn’t scout either of these guys in high school (-0.5) and they sure as heck never played for the Braves (-1.0). Now we’re at 2.0… Screw this, I’m not overpaying for league average players. Let’s just offer Rafael Furcal 5/90 and get it over with.”

-----------------

Rally/6: tremendous

Tango/5, that’s not true.  If just one guy is bidding on a player, you don’t exactly have an efficient market, do you?

I agree that .233 and .305 make these guys FAR from twins, in the eyes of some GMs, and quite possibly in the eyes of Ellis’ agent.

Ok, you need *two* GMs who are as smart as the people who voted in the two polls, and who have a need for an infielder.

I have to believe that the majority of GMs (and their front office people) have a good idea as to the twinnicity of Ellis and Hudson, and will pay scant attention to a one-year batting average.  I will guess there’s at least 20 teams that would value Ellis and Hudson within 0.5 WAR of each other.

Tango/5: ONe more thing, sorry. If their fielding is considered a wash, looking at their wOBA-based bRAA’s over the last 3 seasons could easily lead to the conclusion that Hudson is 1 WAR better.

How can we only need one GM to establish a true value?  If Ellis is truly worth 3/36 (or at least is assumed to be by some GM), and the market dictates that he’s worth 2/12 (or something a shade more than he got with the A’s), I would think any GM smart enough to properly value Ellis is going to bid at somewhere (far?) less than 3/36, possibly a touch over 2/12.

While they might be statistical twins in terms of value, how often do statistical twins with a .070 point difference in average get paid the same? 

Hudson also made 25% more this season, for whatever that’s worth.

In #9, I said two GMs.

***

Devil: Don’t forget that NL players need to be knocked down by 0.5 wins to compare to AL players.  What you are suggesting is that all GMs who are bidding will: a) not account for the league disparity, b) give them a wash on fielding, c) give Hudson +1 win on batting, so that they have a 1 WAR difference.  All that while voters in the poll say they are even.  It’s a stretch, but it’s possible.

From 2006-2008, Orlando Hudson signed a series of one year deals, totalling 12.45 million$.

Mark Ellis signed a 3-year deal (implying a discount) for 2006-08, totalling 11 million$.

11/12.45 is 88%.  And that doesn’t account for the discount that a player gives for signing a multi-year deal.

So, the A’s and DBacks certainly valued them very close to the same.

Tangotiger #12-13: OK… I’m convinced. Thanks. I really wasn’t disagreeing, just needed it spelled out for my feeble little mind. I hadn’t put together my questions with the “just one GM” one, and now it makes more sense.

The As we know about, sort of, but are the Dbacks also considered a “progressive” organization, sabermetrically speaking? I know that some probably only Oakland and Boston to be anywhere close to adequate, but what about ARI, TB, and CLE? I know this comes up fairly often, and I don’t want to take the thread too far off, but I’m wondering how many take this sort of stuff seriously.

As I said in the Ellis thread, I have a fair amount of difference (.5 wins) between the two.

I have both projected as exactly a league average batter.

In defense, here are both of their UZR’s for the last 4 years:

Hudson

05 15
06 4
07 16
08 2

Ellis

05 23
06 12
07 26
08 21

As, you can see, Hudson has a much better defensive rating.

And for the offense, the zero projection is as compared to their respective leagues, so Ellis is better there as well.  In fact, he is probably .75 wins better.

Here are their offensive lwts, again relative to their leagues:

Ellis

05 +24
06 -20
07 +8
08 -8 (not nearly as bad as people think and his .233 BA suggests)

Hudson

05 -19
06 +2
07 +1
08 +8

Of course Hudson is going to be overvalued by virtually ANY team.  One, I think they all think that his defense is better or at least just as good.  Two, his worst year was in 05 and Ellis’ best year was in 05.  No one (but a forecaster) pays any attention to what a player did 3 or 4 years ago. Three, Ellis has a relatively bad year this year and Hudson has his best year.  Fourth, Ellis hit .233 this year.  No one likes a .233 batter (unless maybe they hit 40 or 50 HR - maybe).  Fifth, most teams do not understand how to apply the disparity in talent between the leagues to individual players.

OF COURSE Hudson is going to be valued much higher than Ellis by virtually EVERY team.

"As, you can see, Hudson has a much better defensive rating. “

MGL meant Ellis of course.

MGL meant Ellis of course.

Hey, I may be dumb, but even I figured that one out.

The numbers don’t surprise me, actually, although MGL does seem to agree that teams will overvalue Hudson due to looking at his offensive stats on the surface.

I imagine you guys don’t like PrOPS too much (yuck, OPS!), but it supports MGL’s comments on Ellis. It has him as one of the five unluckiest (qualified) hitters in baseball this season. Sorr if this is redundant, but while we’re at it:

Ellis PrOPS, OPS, OPS-PrOPS

2008: .798, .694, -0.104
2007: .768, .777, .009
2006: .761, .704, -0.057

O-D(o)(aw)g

2008: .780, .817, .036
2007: .783, .817, .033
2006: .784, .809, .024

I suppose that things even out over a number of seasons, though, vis-a-vis projected performance and true talent?

There is still going to be quite a bit of random variance around a player’s true talent, even over a few seasons.  Of course, when you think of all the players in baseball, even a rare 2.5 SD random fluctuation will happen to 6 players or so in any given time period.

Maybe Ellis or his agent watches the stock market and foresees assets of all types (importantly including the value of baseball players) being marked way down.

I don’t like Props much, because it isn’t predictive.  In a very basic way it could be useful, in that a player with a higher props than real production will tend to improve, but we have tested it around here and it’s less predictive than a Marcel. 

Ground balls, Flyballs, and Linedrives among different players are not created equal, and that is the flaw.

Right, I agree with Rally. It’s too dangerous to treat a David Eckstein and Ryan Howard line drive as the same thing.

This is another case of where doing too much is (likely) worse than doing nothing at all.

yes, I always advocate taking something like PrOPS or park factors and erring on the side of caution by assuming and applying a small effect (basically regressing the effect aggressively).  When you do that, you can’t go wrong. You may undershoot the effect (which is OK), but you can’t do worse than nothing at all.  When you use the whole effect, you run the grave risk of doing worse than nothing at all.

Although all of this makes for an interesting situation, I don’t think we would be able to look at Hudson’s contract and say “thats what Ellis should have got.” I say that because now, with Ellis off of the market and Hudson being the only legitimately productive second baseman on the market, the price Hudson will command now is much different than what he would have gotten if Ellis were still available. Anyone follow?
And I agree with you Tom, this move by Ellis makes no sense whatsoever. Money aside, if I were him, and I had no interest in leaving Oakland, I would still want to test free agency just to know what other GM’s think of me as a player. But thats just me.

Isaac, definitely true, although I don’t know that it would be “much different.” Hudson has to be pretty happy that Ellis is off the market though.

Hey, not everyone can be a good negotiator (for lack of a better word). That is the job of an agent actually.  Most players will have no idea what they can or should get.

Either Ellis’s agent screwed up royally or Ellis intervened such that he got what he wanted.  You can’t really “fault” someone for taking 7 mil a year (or whatever it ended up being) when they “could have” (maybe) gotten 10 (or whatever).

He got 2/11, guaranteed.  The TEAM holds the option for the 3rd year.  Talk about bending over.

After Mark Ellis has finished delivering 70 million$ of free agent performance through 2008, at a cost of 7 million$, still the A’s want to hold the option for the third year.  How Beane got the b-lls to ask for that, I have no idea…

One can of course only guess at the reasoning of mr. Beane but hey, his counterpart here is an adult millionaire with professional representation.

I see no particular reason to “hold back” and/or use kid gloves. Since we also don’t know the details of the negotiating process it is certanly possible that parts of this deal were “negtiating points” that Beane and the A’s didn’t expect to hold up.

I.e. “Let’s put in a proposed club option for a 3rd year so we can look generous when we take it of”. Once in a while the oposing party will accept such a thing as an “unexpected bonus”.

And besides, it should not come as a suprise to anyone in baseball that Beane sometimes has a fox or two up his sleeve.

If Beane can rake Ellis and his agent over the coals, all the more power to him.  If I am the owner of the A’s and I thought my GM was giving away my money, just to be a “nice (or supposedly “fair") guy,” I think he would be looking for a new job…

It’s now becoming apparent that Orlando Hudson will sign for Mark Ellis money (2/11, or 3/18).

So, Mark Ellis will have predicted the effect of the collapse of the economy to MLB so well that he volunteered to be paid as if the collapse had already happened.

Hudson ends up at one year, 3.4MM guaranteed and another $4.6MM in incentives. I don’t know if Ellis understood the significance of the financial events in September and I don’t know how risky his health is, but taking 50 cents on the dollar is looking pretty shrewd right now.

If the reason Ellis signed when he did, and for the amount, was because he foresaw the banking events coming up, then I’m sure he made up for the lower contract by shorting the S&P 500.

I was among those who foresaw it, many years in advance, but I had no idea when the first round of ship would hit the fan.  I did OK, avoiding a loss on my assets and turning in a small gain for the year.

is there anything Mark Ellis can’t do? Does this increase his WAR in some way, shape, or fashion?